Quickstart

A quick example to get you started with the Credo AI SDK.

Setup

Python

pip install pycredoai

TypeScript

npm install @credo-ai/sdk

Set your environment variables:

export CREDOAI_API_KEY="your-api-key"
export CREDOAI_TENANT="your-tenant"

Hello World

Python

from credoai import CredoAI, UseCaseCreate

# Initialize the client
client = CredoAI()

# Verify connection
ping = client.system.ping()
print(f"Connected: {ping.message}")

# Create a use case
use_case = client.use_cases.create(
    data=UseCaseCreate(
        name="My First Use Case",
        description="Created via the SDK"
    )
)
print(f"Created: {use_case.name} (ID: {use_case.id})")

# List all use cases
for uc in client.use_cases.list_all():
    print(f"  - {uc.name}")

# Clean up
client.use_cases.delete(use_case_id=use_case.id)
print("Deleted use case")

TypeScript

import { createCredoAIClient } from '@credo-ai/sdk';

// Initialize the client
const client = createCredoAIClient('your-tenant');

// Verify connection
const ping = await client.system.ping();
if (ping.error) {
  console.error('Connection failed:', ping.error.status);
  process.exit(1);
}
console.log('Connected:', ping.data.status);

// Create a use case
const { data: useCase, error } = await client.useCases.create({
  name: 'My First Use Case',
  description: 'Created via the SDK',
});
if (error) throw error;
console.log(`Created: ${useCase.name} (ID: ${useCase.id})`);

// List all use cases
const { data: list } = await client.useCases.list();
if (list) {
  for (const uc of list.items) {
    console.log(`  - ${uc.name}`);
  }
}

// Clean up
await client.useCases.delete(useCase.id);
console.log('Deleted use case');

What's Next?

Now that you have the basics working:

• Learn about Use Cases, Models, and Vendors
• Understand Relationships between resources
• Master Pagination for large datasets
• Handle Errors gracefully
• Use Async Operations for better performance